• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Look for high-voltage regulator projects

Status
Not open for further replies.
Jump to Latest
#142
anatech said:
Hi Jack,

That may contribute to the "sound" of a regulator, so you have asked a valid question. An active regulator is still an amplifier with a reference voltage as the signal.

-Chris
Click to expand...
Chris,
Does the impedance of the power supply contribute to that of the reference voltage, even if it's under a current source? Is the ratio Vin/Vref important?
 
#144
No, it's not clear to me what's causing this drift. The semiconductors don't feel warm. Working around this little inconvenience one can set a somewhat lower target voltage of cause, but I don't think it's intended to work this way.

It might be the 18V zeners employed to set Vgs are the troublemakers as drift is temperature dependend.
 
#145
disco said:

It might be the 18V zeners employed to set Vgs are the troublemakers as drift is temperature dependend.
Click to expand...

Just a side-bar conversation here -- when I started using PbF (lead free) h.v. zeners they seemed to drift much more than the old ones on hand -- maybe it was just the phase of the moon or something? They also failed when run at their rated current/dissipation.

When Motorola discontinued the MC1466 the ham radio community tore their garments, gnashed their teeth and rolled on the ground with foaming mouths...perhaps Mr. Janneman could convince the good offices of EDN to find a replacement for the "obsolete" 2SJ74. (Off soapbox).

R3 and R4 on the output boards should probably be rated for high voltage.

Do not connect this to your '3577.
 
#146
jackinnj said:
R3 and R4 on the output boards should probably be rated for high voltage.

Do not connect this to your '3577.
Click to expand...
Hi Jack,

Just had some measurements in a 'proper' setup 🙂 There's only 3.6 volt over each 18V zener diode, so I guess these are not causing the drift.

You referred to R3 and R4 as in the attached schematic?
What's a '3577?

An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
 
#147
No, I was referring to R3 and R4 as the output resistor's in Jan's 3/2009 Elektor article.

A 3577 is an HP3577 Network Analyzer -- the front end doesn't like D.C.

I like the notion of the Supertex HV Depletion MOSFETs in this application -- I got a bunch for experimentation with SY's IMPASSE preamplifier.
 
#149
jackinnj said:
[snip]When Motorola discontinued the MC1466 the ham radio community tore their garments, gnashed their teeth and rolled on the ground with foaming mouths...
[snip]
Click to expand...

I got a mail from a guy in France mentioning the MC1566L - a later version of the 1466? There is an app with the 1466/1566 on my website that I got after the article. Anyway, I have a source in Germany for 1466's at 15 euro each (that was last year).

Jan Didden
 
#153
Disco, I think the zeners are there purely as protection to ensure the Vgs stays below it's max value. Have you looked at increasing R2 to 56K?

This does 2 things- Increases the load on Q5 for higher gain for error correction and increases the voltage across Q1 improving the operating point of Q1.
 
#154
jkeny said:
Disco, I think the zeners are there purely as protection to ensure the Vgs stays below it's max value.
Click to expand...
Yes, figured that one out ofter the 3.6V reading :~)
Have you looked at increasing R2 to 56K? This does 2 things- Increases the load on Q5 for higher gain for error correction and increases the voltage across Q1 improving the operating point of Q1.
Click to expand...
Tried that one already without succes...
 
#158
Another point is that there can be a wide variety in the LND150 parts so you need about 0.33mA from the Q4/Q3/R1 CCS so:
Current and voltage can be set independently. Current in the bias string is set by R1. The voltage across R2, sets the voltage difference between the gates (and outputs) of Q2 and Q1. This drop across R2 is, of course, set by the current delivered by the Q3/Q4 cascode. So if you decrease R1 to increase the current, you have to decrease R2 by the same proportion, to maintain the same voltage relationships as before.
Click to expand...
 
#160
jkeny said:
Another point is that there can be a wide variety in the LND150 parts so you need about 0.33mA from the Q4/Q3/R1 CCS so:
Click to expand...
Yep, first thing I do when assembling a board (made four so far, all encountering this drift phenomenon) is fidling with R1 so there's 9.00 volt over R2 (27K) which equals 0.33mA through the ref string.
Current and voltage can be set independently. Current in the bias string is set by R1. The voltage across R2, sets the voltage difference between the gates (and outputs) of Q2 and Q1. This drop across R2 is, of course, set by the current delivered by the Q3/Q4 cascode. So if you decrease R1 to increase the current, you have to decrease R2 by the same proportion, to maintain the same voltage relationships as before.
Click to expand...
Hey, this paragraphe is new to me. Where did you find it? What would be the ideal voltage relationship for Q1 and Q2?

Altering the layout made no difference: R4 and R5 are now right at the gates of Q1 and Q2. Neither did upping the voltage. My test situation is: unregulated input varies from 372Vdc to 386Vdc (mains fluctuation), as a result current drawn from the PS 56mA to 58mA, load 274Vdc @54mA steady. It takes about 8 minutes before output stabilises and remains constant over time and over load changes. Rise in output voltage is circa 1.5%.

To be absolutely shure I will have another look at the power dissipation in the resisters.
 
Status
Not open for further replies.
Top Bottom